by Rohit Alluri
In our previous article, we talked about the reasons behind having an “Insurance Fund” and how it actually works. Insurance funds are necessary to protect winning traders against calamitously volatile swings in prices. However, in some extreme cases, capital buffer in insurance funds are not adequate to keep the winning traders’ profits whole, leading to an unsavory scenario in which the winners’ profits are clawed back to compensate for the losses made by the losers.
Let’s consider an example in which winners’ profits are clawed back as losing traders do not have enough capital to cover their losses.
Consider two traders A and B who take opposite positions (one long and one short) with 10x leverage and $1000 in their margin accounts. Assume the price of BTC is $10k at the entry
Let’s consider a scenario in which the price of BTC goes up by 10% to $11000. In which case, Trader B’s position will be taken over by the exchange and will be liquidated as he does not have enough margin to cover his position. In the liquidation process, if the exchange sells the position for a price that is greater than $11000, say $11400, due to price slippage and high market volatility. In that case, the winning trader has made a profit of $1400, but the losing trader has only $1000 in his margin account, leaving the winner shortchanged by $400.
In cases like this, the exchange taps into its insurance fund to keep the winners’ profits whole, but if the insurance fund does not have sufficient funds, winners’ profits are clawed back and limited to what the losing trader has in his margin account. In the above example, Trader A’s are capped at $1000 despite him having made $1400 on his trade. This is called socializing losses in which the losses made by losing traders that are more than the margin deposited are compensated for by the winning traders.
In real life, socializing losses could turn out to be pretty ugly for winners as was the case with OkEx in 2018 when a $416 million position was taken over by the exchange and when the price slippage caused the average selling price for that position to drop below the bankruptcy price, winners had to forego a significant chunk of their profits.
In the night of July 31st 2018, an unprecedentedly large Bitcoin long position got liquidated on Chinese cryptocurrency exchange OKEx. The position had a total value of a staggering $416,000,000 USD (four hundred sixteen million) on entry.
It is unknown what leverage was applied to this trade, but various findings point out that it must have been x20. This means the trader in question actually had a collateral of $20,800,000 in their trading account, and exchange OKEx lended the trader another $395,200,000 in Bitcoin to increase the position to a total value of $416,000,000.
What happened exactly?
The trader got liquidated when the Bitcoin price reached a level of $8,020.49. Assuming a leverage of x20 (will explain later why), this means the position was at a -5% loss here (-5% x 20 = -100% = liquidation). It also points out the average entrypoint of this long position must have been (8,020.49/95 = 84.42621, 84.42621 * 100 = 8442.62) $8442.62. If you now look at the Bitcoin charts during the aforementioned timeframe, you might scratch your head for a while, because the Bitcoin price had not reached this level since May 21st earlier that year, and the position was not open back then. Yet it is speculated that his entrypoint was still $8442.62. This is because the position was not opened in a Bitcoin spot market, but on the Bitcoin future market. On OKEx, the exchange where these future contracts were traded, those contracts had been trading for a premium up to $8,600+ only a few days in prior to this liquidation. One of the reasons that the contracts were trading at such a high price is because the trader in question had to discreetly build out their huge position by buying up bits and pieces over time, slowly pushing the price for the future contracts up because of the demand.
What happens at liquidation?
The Bitcoin price dropped 5%, and the traders’ future contracts are now worth just $395,200,000 (95% of 416,000,000). The personal collateral from the trader’s account has now vaporized in value ($20,800,000), and so OKEx comes in to take over the position and sell everything that is left in order to secure the funds that they lent out. This is where a problem develops. OKEx wants to sell the remaining contracts worth $395,200,000, but the market is by far not liquid enough. There are simply not enough buyers at that price, and so OKEx is stuck with a really, really huge pile of Bitcoin long contracts at a Bitcoin price of $8,020.49. If the Bitcoin price continues to decline from here, OKEx is going to lose value on these contracts (a lot). And so OKEx puts all contracts up for sale at a price of $8,020.49. All these contracts are now sitting in the sell order book of the exchange, waiting for a buyer.
Who covers the losses of the exchange?
The people that were short on the same exchange against all the contracts that were long (also the contracts that OKEx was still sitting on), were making pretty good profits as they made the right bet. But OKEx’s terms and conditions state that in a situation like this, where OKEx loses money because of an unfilled order after liquidation, all counterparties of the liquidated trade will have to cover for these losses by handing over a share of their profits. This is called clawback. In this case, that would have meant that all counterparties (the short positions) would have to hand over almost 50% of their profits. Clawback is not very uncommon, but a clawback of 50% is insane, and therefore unacceptable. OKEx eventually decided to inject 2,500 BTC from their private funds into their insurance fund to cover for a large part of the losses instead. The clients that were short still faced a significant clawback, but already much less severe.